DOCSLIB.ORG

Chapter 1 PREFERENCE MODELLING

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Chapter 1 PREFERENCE MODELLING Chapter 1 PREFERENCE MODELLING Meltem Ozt¨urk,¨ Alexis Tsouki`as LAMSADE-CNRS, Universit´eParis Dauphine, 75775 Paris Cedex 16, France {ozturk,tsoukias}@lamsade.dauphine.fr Philippe Vincke Universit´eLibre de Brussels CP 210/1, Bld. du Triomphe, 1050 Bruxelles, Belgium [email protected] Abstract This chapter provides the reader with a presentation of preference mod- elling fundamental notions as well as some recent results in this field. Preference modelling is an inevitable step in a variety of fields: econ- omy, sociology, psychology, mathematical programming, even medicine, archaeology, and obviously decision analysis. Our notation and some basic definitions, such as those of binary relation, properties and or- dered sets, are presented at the beginning of the chapter. We start by discussing different reasons for constructing a model or preference. We then go through a number of issues that influence the construc- tion of preference models. Different formalisations besides classical logic such as fuzzy sets and non-classical logics become necessary. We then present different types of preference structures reflecting the behavior of a decision-maker: classical, extended and valued ones. It is relevant to have a numerical representation of preferences: functional representa- tions, value functions. The concepts of thresholds and minimal represen- tation are also introduced in this section. In section 7, we briefly explore the concept of deontic logic (logic of preference) and other formalisms associated with “compact representation of preferences” introduced for special purposes. We end the chapter with some concluding remarks. Keywords: Preference modelling, decision aiding, uncertainty, fuzzy sets, non clas- sical logic, ordered relations, binary relations. 1 2 1. Introduction The purpose of this chapter is to present fundamental notions of prefer- ence modelling as well as some recent results in this field. Basic refer- ences on this issue can be considered: [4, 75, 78, 82, 110, 118, 161, 165, 167, 184, 188]. The chapter is organized as follows: The purpose for which formal models of preference and more generally of objects comparison are stud- ied, is introduced in section 1. In section 2, we analyse the information used when such models are established and introduce different sources and types of uncertainty. Our notation and some basic definitions, such as those of binary relation, properties and ordered sets, are presented in section 3. Besides classical logic, different formalisms can be used in order to establish a preference model, such as fuzzy sets and non- classical logics. These are discussed in section 4. In section 5, we then present different types of preference structures reflecting the behavior of a decision-maker: classical, extended and valued ones. It appears relevant to have a numerical representation of preferences: functional representations, value functions and intervals. These are discussed in section 6. The concepts of thresholds and minimal representation are also introduced in this section. Finally, after briefly exploring the con- cept of deontic logic (logic of preference) and other related issued in section 7, we end the chapter with some concluding remarks 2. Purpose Preference modelling is an inevitable step in a variety of fields. Scien- tists build models in order to better understand and to better represent a given situation; such models may also be used for more or less oper- ational purposes (see [30]). It is often the case that it is necessary to compare objects in such models, basically in order to either establish if there is an order between the objects or to establish whether such objects are “near”. Objects can be everything, from candidates to time intervals, from computer codes to medical patterns, from prospects (lot- teries) to production systems. This is the reason why preference mod- elling is used in a great variety of fields such as economy [9, 10, 11, 50], sociology, psychology [37, 42, 45, 112, 111], political science [13, 179], ar- tificial intelligence [65], computer science [82, 177, 188], temporal logic (see [5]) and the interval satisfiability problem [92, 150], mathemati- cal programming [157, 158], electronic business, medicine and biology [22, 38, 108, 114, 138], archaeology [102], and obviously decision analy- sis. Preference Modelling 3 In this chapter, we are going to focus on preference modelling for decision aiding purposes, although the results have a much wider validity. Throughout this chapter, we consider the case of somebody (possibly a decision-maker) who tries to compare objects taking into account dif- ferent points of view. We denote the set of alternatives A1, to be labelled a, b, c, ... and the set of points of view J, labelled j = 1, 2, ..., m. In this framework, a data gj(a) corresponds to the evaluation of the alternative a from the point of view j ∈ J. As already mentioned, comparing two objects can be seen as looking for one of the two following possible situations: Object a is “before” object b, where “before” implies some kind of order between a and b, such an order referring either to a direct preference (a is preferred to b) or being induced from a measure- ment and its associated scale (a occurs before b, a is longer, bigger, more reliable, than b); Object a is “near” object b, where “near” can be considered ei- ther as indifference (object a or object b will do equally well for some purpose), or as a similarity, or again could be induced by a measurement (a occurs simultaneously with b, they have the same length, weight, reliability). The two above-mentioned “attitudes” (see [142]) are not exclusive. They just stand to show what type of problems we focus on. From a decision aiding point of view we traditionally focus on the first sit- uation. Ordering relations is the natural basis for solving ranking or choice problems. The second situation is traditionally associated with problems where the aim is to be able to put together objects sharing a common feature in order to form “homogeneous” classes or categories (a classification problem). The first case we focus on is the ordering relation: given the set A, establishing how each element of A compares to each other element of A from a “preference” point of view enables to obtain an order which might be used to make either a choice on the set A (identify the best) or to rank the set A. Of course, we have to consider whether it is possible to establish such an ordering relation and of what type (certain, uncertain, strong, weak etc.) for all pairs of elements of A. We also have to establish what “not preference” represents (indifference, incomparability etc.). In the following sections (namely in section 5), we are going to see that different options are available, leading to different so called preference structures. In the second case we focus on the “nearness” relation since the is- sue here is to put together objects which ultimately are expected to be 4 “near” (whatever the concept of “near” might represent). In such a case, there is also the problem how to consider objects which are “not near”. Typical situations in this case include the problems of grouping, discriminating and assigning [98]. A further distinction in such problems concerns the fact that the categories with which the objects might be associated could already exist or not and the fact that such categories might be ordered or not. Putting objects into non pre-existing non or- dered categories is the typical classification problem, conversely, assign- ing objects to pre-existing ordered categories is known as the “sorting” problem [149, 154, 220]. It should be noted that although preference relations have been natu- rally associated to ranking and choice problem statements, such a sepa- ration can be argued. For instance, there are sorting procedures (which can be seen as classification problems) that use preference relations in- stead of “nearness” ones [126, 136, 215]. The reason is the following: in order to establish that two objects belong to the same category we usu- ally either try to check whether the two objects are “near” or whether they are near a “typical” object of the category (see for instance [154]). If, however, a category is described, not through its typical objects, but through its boundaries, then, in order to establish if an object belongs to such a category it might make sense to check whether such an ob- ject performs “better” than the “minimum”, or “least” boundary of the category and that will introduce the use of a preference relation. Recently Ngo The [142] claimed that decision aiding should not ex- clusively focus on preference relations, but also on “nearness relations”, since quite often the problem statement to work with in a problem formu- lation is that of classification (on the existence of different problem state- ments and their meaning the reader is referred to [172, 173, 52, 204]). 3. Nature of Information As already mentioned, the purpose of our analysis is to present the literature associated with objects comparison for either a preference or a nearness relation. Nevertheless, such an operation is not always as intuitive as it might appear. Building up a model from reality is always an abstraction (see [28]). This can always be affected by the presence of uncertainty due to our imperfect knowledge of the world, our limited capability of observation and/or discrimination, the inevitable errors occurring in any human activity etc. [170]. We call such an uncertainty exogenous. Besides, such an activity might generate uncertainty since Preference Modelling 5 it creates an approximation of reality, thus concealing some features of reality. We call this an endogenous uncertainty (see [190]). As pointed out by Vincke [205] preference modelling can be seen as either the result of direct comparison (asking a decision-maker to com- pare two objects and to establish the relation between them) from which it might be possible to infer a numerical representation, or as the result of the induction of a preference relation from the knowledge of some “measures” associated to the compared objects.
Load more

Recommended publications
  • Math 475 Homework #3 March 1, 2010 Section 4.6
    Student: Yu Cheng (Jade) Math 475 Homework #3 March 1, 2010 Section 4.6 Exercise 36-a Let ͒ be a set of ͢ elements. How many different relations on ͒ are there? Answer: On set ͒ with ͢ elements, we have the following facts. ) Number of two different element pairs ƳͦƷ Number of relations on two different elements ) ʚ͕, ͖ʛ ∈ ͌, ʚ͖, ͕ʛ ∈ ͌ 2 Ɛ ƳͦƷ Number of relations including the reflexive ones ) ʚ͕, ͕ʛ ∈ ͌ 2 Ɛ ƳͦƷ ƍ ͢ ġ Number of ways to select these relations to form a relation on ͒ 2ͦƐƳvƷͮ) ͦƐ)! ġ ͮ) ʚ ʛ v 2ͦƐƳvƷͮ) Ɣ 2ʚ)ͯͦʛ!Ɛͦ Ɣ 2) )ͯͥ ͮ) Ɣ 2) . b. How many of these relations are reflexive? Answer: We still have ) number of relations on element pairs to choose from, but we have to 2 Ɛ ƳͦƷ ƍ ͢ include the reflexive one, ʚ͕, ͕ʛ ∈ ͌. There are ͢ relations of this kind. Therefore there are ͦƐ)! ġ ġ ʚ ʛ 2ƳͦƐƳvƷͮ)Ʒͯ) Ɣ 2ͦƐƳvƷ Ɣ 2ʚ)ͯͦʛ!Ɛͦ Ɣ 2) )ͯͥ . c. How many of these relations are symmetric? Answer: To select only the symmetric relations on set ͒ with ͢ elements, we have the following facts. ) Number of symmetric relation pairs between two elements ƳͦƷ Number of relations including the reflexive ones ) ʚ͕, ͕ʛ ∈ ͌ ƳͦƷ ƍ ͢ ġ Number of ways to select these relations to form a relation on ͒ 2ƳvƷͮ) )! )ʚ)ͯͥʛ )ʚ)ͮͥʛ ġ ͮ) ͮ) 2ƳvƷͮ) Ɣ 2ʚ)ͯͦʛ!Ɛͦ Ɣ 2 ͦ Ɣ 2 ͦ . d.
    [Show full text]
  • Extending Partial Isomorphisms of Finite Graphs
    S´eminaire de Master 2`emeann´ee Sous la direction de Thomas Blossier C´edricMilliet — 1er d´ecembre 2004 Extending partial isomorphisms of finite graphs Abstract : This paper aims at proving a theorem given in [4] by Hrushovski concerning finite graphs, and gives a generalization (with- out any proof) of this theorem for a finite structure in a finite relational language. Hrushovski’s therorem is the following : any finite graph G embeds in a finite supergraph H so that any local isomorphism of G extends to an automorphism of H. 1 Introduction In this paper, we call finite graph (G, R) any finite structure G with one binary symmetric reflexive relation R (that is ∀x ∈ G xRx and ∀x, y xRy =⇒ yRx). We call vertex of such a graph any point of G, and edge, any couple (x, y) such that xRy. Geometrically, a finite graph (G, R) is simply a finite set of points, some of them being linked by edges (see picture 1 ). A subgraph (F, R0) of (G, R) is any subset F of G along with the binary relation R0 induced by R on F . H H H J H J H J H J Hqqq HJ Hqqq HJ ¨ ¨ q q ¨ q q q q q Picture 1 — A graph (G, R) and a subgraph (F, R0) of (G, R). We call isomorphism between two graphs (G, R) and (G0,R0) any bijection that preserves the binary relations, that is, any bijection σ that sends an edge on an edge along with σ−1. If (G, R) = (G0,R0), then such a σ is called an automorphism of (G, R).
    [Show full text]
  • PROBLEM SET THREE: RELATIONS and FUNCTIONS Problem 1
    PROBLEM SET THREE: RELATIONS AND FUNCTIONS Problem 1 a. Prove that the composition of two bijections is a bijection. b. Prove that the inverse of a bijection is a bijection. c. Let U be a set and R the binary relation on ℘(U) such that, for any two subsets of U, A and B, ARB iff there is a bijection from A to B. Prove that R is an equivalence relation. Problem 2 Let A be a fixed set. In this question “relation” means “binary relation on A.” Prove that: a. The intersection of two transitive relations is a transitive relation. b. The intersection of two symmetric relations is a symmetric relation, c. The intersection of two reflexive relations is a reflexive relation. d. The intersection of two equivalence relations is an equivalence relation. Problem 3 Background. For any binary relation R on a set A, the symmetric interior of R, written Sym(R), is defined to be the relation R ∩ R−1. For example, if R is the relation that holds between a pair of people when the first respects the other, then Sym(R) is the relation of mutual respect. Another example: if R is the entailment relation on propositions (the meanings expressed by utterances of declarative sentences), then the symmetric interior is truth- conditional equivalence. Prove that the symmetric interior of a preorder is an equivalence relation. Problem 4 Background. If v is a preorder, then Sym(v) is called the equivalence rela- tion induced by v and written ≡v, or just ≡ if it’s clear from the context which preorder is under discussion.
    [Show full text]
  • Mereology Then and Now
    Logic and Logical Philosophy Volume 24 (2015), 409–427 DOI: 10.12775/LLP.2015.024 Rafał Gruszczyński Achille C. Varzi MEREOLOGY THEN AND NOW Abstract. This paper offers a critical reconstruction of the motivations that led to the development of mereology as we know it today, along with a brief description of some questions that define current research in the field. Keywords: mereology; parthood; formal ontology; foundations of mathe- matics 1. Introduction Understood as a general theory of parts and wholes, mereology has a long history that can be traced back to the early days of philosophy. As a formal theory of the part-whole relation or rather, as a theory of the relations of part to whole and of part to part within a whole it is relatively recent and came to us mainly through the writings of Edmund Husserl and Stanisław Leśniewski. The former were part of a larger project aimed at the development of a general framework for formal ontology; the latter were inspired by a desire to provide a nominalistically acceptable alternative to set theory as a foundation for mathematics. (The name itself, ‘mereology’ after the Greek word ‘µρoς’, ‘part’ was coined by Leśniewski [31].) As it turns out, both sorts of motivation failed to quite live up to expectations. Yet mereology survived as a theory in its own right and continued to flourish, often in unexpected ways. Indeed, it is not an exaggeration to say that today mereology is a central and powerful area of research in philosophy and philosophical logic. It may be helpful, therefore, to take stock and reconsider its origins.
    [Show full text]
  • Section 4.1 Relations
    Binary Relations (Donny, Mary) (cousins, brother and sister, or whatever) Section 4.1 Relations - to distinguish certain ordered pairs of objects from other ordered pairs because the components of the distinguished pairs satisfy some relationship that the components of the other pairs do not. 1 2 The Cartesian product of a set S with itself, S x S or S2, is the set e.g. Let S = {1, 2, 4}. of all ordered pairs of elements of S. On the set S x S = {(1, 1), (1, 2), (1, 4), (2, 1), (2, 2), Let S = {1, 2, 3}; then (2, 4), (4, 1), (4, 2), (4, 4)} S x S = {(1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2, 3), (3, 1), (3, 2) , (3, 3)} For relationship of equality, then (1, 1), (2, 2), (3, 3) would be the A binary relation can be defined by: distinguished elements of S x S, that is, the only ordered pairs whose components are equal. x y x = y 1. Describing the relation x y if and only if x = y/2 x y x < y/2 For relationship of one number being less than another, we Thus (1, 2) and (2, 4) satisfy . would choose (1, 2), (1, 3), and (2, 3) as the distinguished ordered pairs of S x S. x y x < y 2. Specifying a subset of S x S {(1, 2), (2, 4)} is the set of ordered pairs satisfying The notation x y indicates that the ordered pair (x, y) satisfies a relation .
    [Show full text]
  • Relations II
    CS 441 Discrete Mathematics for CS Lecture 22 Relations II Milos Hauskrecht [email protected] 5329 Sennott Square CS 441 Discrete mathematics for CS M. Hauskrecht Cartesian product (review) •Let A={a1, a2, ..ak} and B={b1,b2,..bm}. • The Cartesian product A x B is defined by a set of pairs {(a1 b1), (a1, b2), … (a1, bm), …, (ak,bm)}. Example: Let A={a,b,c} and B={1 2 3}. What is AxB? AxB = {(a,1),(a,2),(a,3),(b,1),(b,2),(b,3)} CS 441 Discrete mathematics for CS M. Hauskrecht 1 Binary relation Definition: Let A and B be sets. A binary relation from A to B is a subset of a Cartesian product A x B. Example: Let A={a,b,c} and B={1,2,3}. • R={(a,1),(b,2),(c,2)} is an example of a relation from A to B. CS 441 Discrete mathematics for CS M. Hauskrecht Representing binary relations • We can graphically represent a binary relation R as follows: •if a R b then draw an arrow from a to b. a b Example: • Let A = {0, 1, 2}, B = {u,v} and R = { (0,u), (0,v), (1,v), (2,u) } •Note: R A x B. • Graph: 2 0 u v 1 CS 441 Discrete mathematics for CS M. Hauskrecht 2 Representing binary relations • We can represent a binary relation R by a table showing (marking) the ordered pairs of R. Example: • Let A = {0, 1, 2}, B = {u,v} and R = { (0,u), (0,v), (1,v), (2,u) } • Table: R | u v or R | u v 0 | x x 0 | 1 1 1 | x 1 | 0 1 2 | x 2 | 1 0 CS 441 Discrete mathematics for CS M.
    [Show full text]
  • Chарtеr 2 RELATIONS
    Ch=FtAr 2 RELATIONS 2.0 INTRODUCTION Every day we deal with relationships such as those between a business and its telephone number, an employee and his or her work, a person and other person, and so on. Relationships such as that between a program and a variable it uses and that between a computer language and a valid statement in this language often arise in computer science. The relationship between the elements of the sets is represented by a structure, called relation, which is just a subset of the Cartesian product of the sets. Relations are used to solve many problems such as determining which pairs of cities are linked by airline flights in a network or producing a useful way to store information in computer databases. In this chapter, we will study equivalence relation, equivalence class, composition of relations, matrix of relations, and closure of relations. 2.1 RELATION A relation is a set of ordered pairs. Let A and B be two sets. Then a relation from A to B is a subset of A × B. Symbolically, R is a relation from A to B iff R Í A × B. If (x, y) Î R, then we can express it by writing xRy and say that x is related to y with relation R. Thus, (x, y) Î R Û xRy 2.2 RELATION ON A SET A relation R on a set A is the subset of A × A, i.e., R Í A × A. Here both the sets A and B are same. -xample Let A = {2, 3, 4, 5} and B = {2, 4, 6, 10, 12}.
    [Show full text]
  • Functions, Relations, Partial Order Relations Definition: the Cartesian Product of Two Sets a and B (Also Called the Product
    Functions, Relations, Partial Order Relations Definition: The Cartesian product of two sets A and B (also called the product set, set direct product, or cross product) is defined to be the set of all pairs (a,b) where a ∈ A and b ∈ B . It is denoted A X B. Example : A ={1, 2), B= { a, b} A X B = { (1, a), (1, b), (2, a), (2, b)} Solve the following: X = {a, c} and Y = {a, b, e, f}. Write down the elements of: (a) X × Y (b) Y × X (c) X 2 (= X × X) (d) What could you say about two sets A and B if A × B = B × A? Definition: A function is a subset of the Cartesian product.A relation is any subset of a Cartesian product. For instance, a subset of , called a "binary relation from to ," is a collection of ordered pairs with first components from and second components from , and, in particular, a subset of is called a "relation on ." For a binary relation , one often writes to mean that is in . If (a, b) ∈ R × R , we write x R y and say that x is in relation with y. Exercise 2: A chess board’s 8 rows are labeled 1 to 8, and its 8 columns a to h. Each square of the board is described by the ordered pair (column letter, row number). (a) A knight is positioned at (d, 3). Write down its possible positions after a single move of the knight. Answer: (b) If R = {1, 2, ..., 8}, C = {a, b, ..., h}, and P = {coordinates of all squares on the chess board}, use set notation to express P in terms of R and C.
    [Show full text]
  • A Method of Finding Automorphism Groups of Endomorphism Monoids
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector Discrete Mathematics 307 (2007) 1609–1620 www.elsevier.com/locate/disc A method of finding automorphism groups of endomorphism monoids of relational systems João Araújoa,b, Janusz Koniecznyc aUniversidade Aberta, R. Escola Politécnica, 147, 1269-001 Lisboa, Portugal bCentro de Álgebra, Universidade de Lisboa, 1649-003 Lisboa, Portugal cDepartment of Mathematics, University of Mary Washington, Fredericksburg, VA 22401, USA Received 21 September 2004; received in revised form 16 August 2006; accepted 2 September 2006 Available online 16 November 2006 Abstract For any set X and any relation on X, let T(X,) be the semigroup of all maps a : X → X that preserve . Let S(X) be the symmetric group on X.If is reflexive, the group of automorphisms of T(X,) is isomorphic to NS(X)(T (X, )), the normalizer of T(X,) in S(X), that is, the group of permutations on X that preserve T(X,) under conjugation. The elements of NS(X)(T (X, )) have been described for the class of so-called dense relations . The paper is dedicated to applications of this result. © 2006 Elsevier B.V. All rights reserved. MSC: 20M20; 20M15; 05C25; 05C65 Keywords: Automorphism group; Endomorphism; Transformation semigroup; Reflexive relation 1. Introduction For a mathematical structure A, let End(A) denote the monoid of endomorphisms of A. A large amount of effort in mathematical research has been devoted to the study of relations between the monoid End(A) and the structure A itself.
    [Show full text]
  • Introduction to Relations Binary Relation
    Introduction to Relations CSE 191, Class Note 09 Computer Sci & Eng Dept SUNY Buffalo c Xin He (University at Buffalo) CSE 191 Descrete Structures 1 / 57 Binary relation Definition: Let A and B be two sets. A binary relation from A to B is a subset of A B. × In other words, a binary relation from A to B is a set of pair (a, b) such that a A and b B. ∈ ∈ We often omit “binary” if there is no confusion. If R is a relation from A to B and (a, b) R, we say a is related to b by R. ∈ We write a R b. Example: Let A be the set of UB students, and B be the set of UB courses. Define relation R from A to B as follows: a R b if and only if student a takes course b. So for every student x in this classroom, we have that (x, CSE191) R, or ∈ equivalently, x R CSE191. For your TA Ding , we know that Ding is NOT related to CSE191 by R. c XinSo He (University(Ding, CSE at Buffalo)191) R. CSE 191 Descrete Structures 3 / 57 6∈ Relation on a set We are particularly interested in binary relations from a set to the same set. If R is a relation from A to A, then we say R is a relation on set A. Example: We can define a relation R on the set of positive integers such that a R b if and only if a b. | 3 R 6.
    [Show full text]
  • Binary Relations.Pdf
    Binary Relations Any set of ordered pairs defines a binary relation. We express a particular ordered pair, (x, y) R, where R is a binary relation, as xRy. Properties Properties of a binary relation R on a set X: a. reflexive: if for every x X, xRx holds, i.e. (x, x) R. b. symmetric: if for every x and y in X, whenever xRy holds, then yRx holds. c. transitive: if for every x, y, and z in X, whenever xRy holds and yRz holds, then xRz holds. Graphical Representation Suppose that X = {A, B, C, D} and that R = {(A, B), (B, D), (C, C), (D, B)}. We can represent the binary relation R as a set, as a graph or as a Boolean matrix. Graphically, we have Boolean Matrix Representation As a Boolean matrix we have A B C D A 0 1 0 0 B 0 0 0 1 C 0 0 1 0 D 0 1 0 0 What would a Boolean matrix look like if it represented a reflexive binary relation, a symmetric binary relation, a transitive binary relation, any combination of the above? Reflexive Relation A B C D A - 1 - - B - - - 1 C - - 1 - D - 1 - - Symmetric Relation A B C D A - 1 - - B - - - 1 C - - 1 - D - 1 - - Transitive Relation A B C D A - 1 - - B - - - 1 C - - 1 - D - 1 - - Click here for Answers Composition of Relations: Let R be a relation from X to Y, and S be a relation from Y to Z. Then a relation written as R o S is called a composite relation of R and S where R o S = { (x, z) | x X z Z ( y) (y Y (x, y) R (y, z) S) } We can also write the composition as R o S = { (x, z) | x X z Z ( y) (y Y xRy ySz) } Note: Relational composition can be realized as matrix multiplication.
    [Show full text]
  • Topology Vs. Generalized Rough Sets
    International Journal of Approximate Reasoning 52 (2011) 231–239 Contents lists available at ScienceDirect International Journal of Approximate Reasoning journal homepage: www.elsevier.com/locate/ijar Topology vs generalized rough sets ⇑ Zhi Pei a, Daowu Pei b, , Li Zheng a a Department of Industrial Engineering, Tsinghua University, Beijing 100084, China b Department of Mathematics, Zhejiang Sci-Tech University, Hangzhou 310018, China article info abstract Article history: This paper investigates the relationship between topology and generalized rough sets Received 16 December 2007 induced by binary relations. Some known results regarding the relation based rough sets Received in revised form 18 June 2010 are reviewed, and some new results are given. Particularly, the relationship between differ- Accepted 30 July 2010 ent topologies corresponding to the same rough set model is examined. These generalized Available online 19 August 2010 rough sets are induced by inverse serial relations, reflexive relations and pre-order rela- tions, respectively. We point that inverse serial relations are weakest relations which can Keywords: induce topological spaces, and that different relation based generalized rough set models Rough set theory will induce different topological spaces. We proved that two known topologies correspond- Generalized rough set Relation based rough set ing to reflexive relation based rough set model given recently are different, and gave a Topology condition under which the both are the same topology. Ó 2010 Elsevier Inc. All rights reserved. 1. Introduction In past several years, rough set theory (see [15],or[25]) has developed significantly due to its wide applications. Various generalized rough set models have been established and their properties or structures have been investigated intensively (see [36,37],or[35,20–23,11]).
    [Show full text]